论文检索

Deep Learning · Generative Models and Autoencoders

Kaisen Yang, Jayden Teoh, Kaicheng Yang, Yitong Zhang, Alex Lamb

Masked Diffusion Models (MDMs) offer greater flexibility in decoding order than autoregressive models but require careful path planning to achieve high-quality generation. While existing samplers typically greedily select positions with the lowest uncertainty, we identify their fundamental limitations through failure case analysis, showing they overlook the impact of current actions on subsequent steps and fail to optimize cumulative uncertainty. To bridge this gap, we propose the **Info-Gain Sampler**, a principled decoding framework that balances immediate costs with information gain. Our method features a simple but effective objective and an efficient implementation that ensures practical overhead is minimal. Extensive evaluations across diverse architectures and tasks (reasoning, coding, creative writing, and image generation) demonstrate that Info-Gain Sampler consistently outperforms existing samplers, significantly raising the performance ceiling of MDMs. For instance, it achieves a 5.5\% improvement in average accuracy on reasoning tasks and a 63.1\% win-rate in creative writing; notably, on reasoning tasks it reduces cumulative uncertainty from 78.4 to 48.6, outperforming the best baseline by a large margin.

Theory · Probabilistic Methods

Yaniv Tenzer, Elad Tolochinksy, Yaniv Romano

We introduce a testing-by-betting framework that leverages predictions on unlabeled data to enhance the power of sequential hypothesis testing. Given limited samples from the joint distribution of $(X,Y)$, and additional unlabeled samples from the marginal of $X$, we ask how unlabeled data can be used to hypothesize about the distribution of $Y$, and the conditional distribution of $Y\mid X$. We introduce an e-statistic and use it to construct a sequential test. Under standard distributional assumptions---label shift or concept shift---we establish that the test is anytime valid. Furthermore, we show that for binary data, the e-statistic has non-trivial power. Crucially, our approach retains these properties even when the underlying predictions are inaccurate. Through simulations and applications to large language models evaluation, we demonstrate power gains over baseline approaches, including prediction-powered inference. These gains persist even with relatively limited unlabeled data and when predictions have low accuracy due to weak correlation between $X$ and $Y$.

Deep Learning · Large Language Models

Shyam Sundhar Ramesh, Xiaotong Ji, Matthieu Zimmer, Sangwoong Yoon, Zhiyong Wang, Haitham Bou Ammar, Aurelien Lucchi, Ilija Bogunovic

RL-based post-training with GRPO is widely used to improve large language models on individual reasoning tasks. However, real-world deployment requires reliable performance across diverse tasks. A straightforward multi-task adaptation of GRPO often leads to imbalanced outcomes, with some tasks dominating optimization while others stagnate. Moreover, tasks can vary widely in how frequently prompts yield zero advantages (and thus zero gradients), which further distorts their effective contribution to the optimization signal. To address these issues, we propose a novel Multi-Task GRPO (MT-GRPO) algorithm that (i) dynamically adapts task weights to explicitly optimize worst-task performance and promote balanced progress across tasks, and (ii) introduces a ratio-preserving sampler to ensure task-wise policy gradients reflect the adapted weights. Experiments on both 3-task and 9-task settings show that MT-GRPO consistently outperforms baselines in worst-task accuracy. In particular, MT-GRPO achieves 16–28\% and 6\% absolute improvement on worst-task performance over standard GRPO and DAPO, respectively, while maintaining competitive average accuracy. Moreover, MT-GRPO requires 50\% fewer training steps to reach 50\% worst-task accuracy in the 3-task setting, demonstrating substantially improved efficiency in achieving reliable performance across tasks.

Deep Learning · Foundation Models

Yulu Gan, Phillip Isola

Pretraining produces a learned parameter vector that is typically treated as a starting point for further iterative adaptation. In this work, we instead view the outcome of pretraining as a distribution over parameter vectors, whose support already contains task-specific experts. We show that in smaller or insufficiently trained models such expert solutions occupy a negligible fraction of the volume of this distribution, making their discovery reliant on structured optimization methods such as gradient descent. In contrast, in large, well-pretrained models the density of task-experts increases dramatically, so that diverse specialists populate a substantial fraction of the neighborhood around the pretrained weights. Motivated by this perspective, we explore a simple, fully parallel post-training method that samples $N$ parameter vectors at random, selects the top $K$, and ensembles them via majority vote to combine complementary expertise. Despite its simplicity, this approach is competitive with standard post-training methods such as PPO, GRPO, and ES for contemporary large-scale models.

Applications · Everything Else

Xinyu Liu, Zixuan Xie, Amir Moeini, Claire Chen, Shuze Liu, Yu Meng, Aidong Zhang, Shangtong Zhang

While the ecosystem of Lean and Mathlib has enjoyed celebrated success in formal mathematical reasoning with the help of large language models (LLMs), the absence of many folklore lemmas in Mathlib remains a persistent barrier that limits Lean's usability as an everyday tool for mathematicians like \LaTeX{} or Maple. To address this, we introduce \textsc{MathlibLemma}, the first LLM-based multi-agent system to automate the discovery and formalization of mathematical folklore lemmas. This framework constitutes our primary contribution, proactively mining the missing connective tissue of mathematics. Its efficacy is demonstrated by the production of a verified library of folklore lemmas, a subset of which has already been formally merged into the latest build of Mathlib, thereby validating the system's real-world utility and alignment with expert standards. Leveraging this pipeline, we further construct the \textsc{MathlibLemma} benchmark, a suite of approximately 4,000 type-checked Lean statements spanning a broad range of mathematical domains. By transforming the role of LLMs from passive consumers to active contributors, this work establishes a constructive methodology for the self-evolution of formal mathematical libraries.

Deep Learning · Foundation Models

Ting Huang, Zeyu Zhang, Hao Tang

This position paper argues that reinforcement learning (RL) should be used to *adjust* foundation models after pretraining and cold-start supervision, not *abused* as a default recipe for capability creation or early-stage training. We view RL as a high-cost, high-leverage post-training operator that reallocates probability mass toward behaviors a model can already express, but rarely creates new reasoning capacities from scratch in a compute-efficient, stable, and controllable way. This distinction matters now because “RL-zero” narratives risk normalizing expensive and brittle RL-first pipelines as the primary path to reasoning, even though practice increasingly shows that cold-start supervision is a prerequisite for reliable RL and that RL is most effective as targeted refinement. Across modalities and domains, we emphasize a recurring regularity: supervision establishes usable reasoning structure, while RL mainly sharpens correctness, consistency, and constraint satisfaction, especially under hard constraints or distribution shift. We further argue for reward minimalism: simple, verifiable rewards often suffice and reduce proxy-driven failure modes relative to over-engineered reward models. Finally, we discuss how self-supervised RL can support self-evolution when grounded in verifiable signals and structured interaction environments. Together, these arguments motivate treating RL as a disciplined adjustment stage with explicit entry criteria and compute-accountable evaluation.

Deep Learning · Large Language Models

Xi Yu, Dmitrii Torbunov, Soumyajit Mandal, Yihui Ren

The design of Analog and Mixed-Signal (AMS) integrated circuits remains heavily reliant on expert knowledge, with transistor sizing a major bottleneck due to nonlinear behavior, high-dimensional design spaces, and strict performance constraints. Existing Electronic Design Automation (EDA) methods typically frame sizing as static black-box optimization, resulting in inefficient and less robust solutions. Although Large Language Models (LLMs) exhibit strong reasoning abilities, they are not suited for precise numerical optimization in AMS sizing. To address this gap, we propose \textsc{AutoSizer}, a reflective LLM-driven meta-optimization framework that unifies circuit understanding, adaptive search-space construction, and optimization orchestration in a closed loop. It employs a two-loop optimization framework, with an inner loop for circuit sizing and an outer loop that analyzes optimization dynamics and constraints to iteratively refine the search space from simulation feedback. We further introduce \textsc{AMS-SizingBench}, an open benchmark comprising 24 diverse AMS circuits in SKY130 CMOS technology, designed to evaluate adaptive optimization policies under realistic simulator-based constraints. \textsc{AutoSizer} experimentally achieves higher solution quality, faster convergence, and higher success rate across varying circuit difficulties, outperforming both traditional optimization methods and existing LLM-based agents.

Applications · Language, Speech and Dialog

Amir Ivry, Shinji Watanabe

Spoken dialogues with and between voice agents are becoming increasingly common, yet assessing them for their socially harmful content such as violence, harassment, and hate remains text-centric and fails to account for audio-specific cues and transcription errors. We present LALM-as-a-Judge, the first controlled benchmark and systematic study of large audio-language models (LALMs) as safety judges for multi-turn spoken dialogues. We generate 24,000 unsafe and synthetic spoken dialogues in English that consist of 3-10 turns, by having a single dialogue turn including content with one of 8 harmful categories (e.g., violence) and on one of 5 grades, from very mild to severe. On 160 dialogues, 5 human raters confirmed reliable unsafe detection and a meaningful severity scale. We benchmark three open-source LALMs: Qwen2-Audio, Audio Flamingo 3, and MERaLiON as zero-shot judges that output a scalar safety score in $\left[0,1\right]$ across audio-only, transcription-only, or multimodal inputs, along with a transcription-only LLaMA baseline. We measure the judges' sensitivity to detecting unsafe content, the specificity in ordering severity levels, and the stability of the score in dialogue turns. Results reveal architecture- and modality-dependent trade-offs: the most sensitive judge is also the least stable across turns, while stable configurations sacrifice detection of mild harmful content. Transcription quality is a key bottleneck: Whisper-Large may significantly reduce sensitivity for transcription-only modes, while largely preserving severity ordering. Audio becomes crucial when paralinguistic cues or transcription fidelity are category-critical. We summarize all findings and provide actionable guidance for practitioners.

Applications · Neuroscience, Cognitive Science

Timothy Kim, Ulises Obilinovic, Yiliu Wang, Eric SheaBrown, Uygar Sümbül

Connectivity structure shapes neural computation, but inferring this structure from population recordings is degenerate: multiple connectivity structures can generate identical dynamics. Recent work uses low-rank recurrent neural networks (lrRNNs) to infer low-dimensional latent dynamics and connectivity structure from observed activity, enabling a mechanistic interpretation of the dynamics. However, standard approaches for training lrRNNs can recover spurious structures irrelevant to the underlying dynamics. We first characterize the identifiability of connectivity structures in lrRNNs and determine conditions under which a unique solution exists. Then, to find such solutions, we develop an inference framework based on maximum entropy and continuous normalizing flows (CNFs), trained via flow matching. Instead of estimating a single connectivity matrix, our method learns the maximally unbiased distribution over connection weights consistent with observed dynamics. This approach captures complex yet necessary distributions such as heavy-tailed connectivity found in empirical data. We validate our method on synthetic datasets with connectivity structures that generate multistable attractors, limit cycles, and ring attractors, and demonstrate its applicability in recordings from rat frontal cortex during decision-making. Our framework shifts circuit inference from recovering connectivity to identifying which connectivity structures are computationally required, and which are artifacts of underconstrained inference.

Yu Feng, Zhen Tian, Haoran Luo, Xie Yu, Diancheng Cheng, Haoyue Zheng, Shuai Lyu, Ping Zong, Lianyuan Li, xin ge 等

Domain Incremental Learning is a critical scenario that requires models to continuously adapt to new data domains without retraining. However, domain shifts often cause severe performance degradation. To address this, we propose Hybrid Energy-Distance Prompt, a domain-incremental framework inspired by Helmholtz free energy. HEDP introduces an energy regularization loss to enhance the separability of domain representations and a hybrid energy-distance weighted mechanism that fuses energy-based and distance-based cues to improve domain selection and generalization. Experiments on multiple benchmarks, including CORe50, show that HEDP achieves superior performance on unseen domains with a 2.57\% accuracy gain, effectively mitigating catastrophic forgetting and enhancing open-world adaptability. Our code is \href{https://anonymous.4open.science/r/HEDP-C879/}{available here}.

Deep Learning · Large Language Models

Santiago Acevedo, Alessandro Laio, Marco Baroni

We study how syntactic and semantic information is encoded in inner layer representations of Large Language Models (LLMs), focusing on the very large DeepSeek-V3. We find that, by averaging hidden-representation vectors of sentences sharing syntactic structure or meaning, we obtain vectors that capture a significant proportion of the syntactic and semantic information contained in the representations. In particular, subtracting these syntactic and semantic ``centroids'' from sentence vectors strongly affects their similarity with syntactically and semantically matched sentences, respectively, suggesting that syntax and semantics are, at least partially, linearly encoded. We also find that the cross-layer encoding profiles of syntax and semantics are different, and that the two signals can to some extent be decoupled, suggesting differential encoding of these two types of linguistic information in LLM representations.

Deep Learning · Large Language Models

Gaotang Li, Ruizhong Qiu, Xiusi Chen, Heng Ji, Hanghang Tong

Supervised fine-tuning (SFT) is the standard approach for post-training large language models (LLMs), yet it often shows limited generalization. We trace this limitation to its default training objective: negative log likelihood (NLL). While NLL is classically optimal when training from scratch, post-training operates in a different paradigm and could violate its optimality assumptions, where models already encode task-relevant priors and supervision can be long and noisy. Rather than proposing a single universally superior replacement loss, we systematically study various probability-based objectives and characterize when and why different objectives succeed or fail under varying conditions. Through comprehensive experiments and extensive ablation studies across 8 model backbones, 27 benchmarks, and 7 domains, we uncover a critical dimension that governs objective behavior: the model-capability continuum. Near the model-strong end, prior-leaning objectives that downweight low-probability tokens (e.g., $-p$, $-p^{10}$, thresholded variants) consistently outperform NLL; toward the model-weak end, NLL dominates; in between, no single objective prevails. Our theoretical analysis further elucidates how objectives trade places across the continuum, providing a principled foundation for adapting objectives to model capability.

Deep Learning · Other Representation Learning

Jae-Jun Lee, Sung Whan Yoon

A surge of recent advancements has consistently highlighted the superiority of multimodal learning over unimodal approaches across a variety of tasks. However, the theoretical foundations elucidating this advantage remain underexplored: existing theoretical analyses are often constrained by tight assumptions, and lack empirical validation. In this paper, we link this gap by proposing a novel theoretical framework grounded in \textit{convolutional smoothing}, offering a new perspective on how multimodal learning contributes to a smoother loss landscape compared to unimodal learning. Building upon this theoretical foundation, we introduce a simple yet effective distributional training approach based on stochastic modality pairing instead of fixed pairing; thus, further promoting flatter landscape via convolutional smoothing. Our empirical results across various multimodal datasets demonstrate that multimodal models not only achieve better performance but also exhibit smoother loss landscape, which represent better robustness and generalization.

Applications · Everything Else

Yao Lai, Xuyuan Xiong, Zeyue Xue, Guojin Chen, Jing Wang, Xihui Liu, Rui Zhang, Robert Mullins, Bei Yu, Ping Luo

In semiconductor manufacturing, lithography projects circuit layouts onto silicon wafers through an optical mask. As circuit features shrink below the wavelength of light, optical diffraction causes the printed patterns to deviate from their intended layouts. Inverse Lithography Technology (ILT) addresses this challenge by generating optimized masks that enhance the fidelity of pattern transfer onto wafers. While ILT resembles an image synthesis task, its reliance on explicit physical metrics for mask evaluation limits the applicability of existing generative models. We introduce LithoGRPO, an ILT framework that integrates the flow‑matching paradigm with GRPO‑based reinforcement learning (RL) fine‑tuning, enabling efficient exploration of diverse masks for a given target layout. Unlike purely generative or optimization‑based approaches, RL in LithoGRPO exploits the explicitly defined, physics‑based reward function of ILT, enabling optimization under complex, process‑aware constraints. To the best of our knowledge, this is the first framework that unifies flow matching and RL for mask optimization. To improve RL sampling efficiency, we propose a fast shot-counting algorithm for manufacturability evaluation, achieving over 130× speedup while preserving the mask ranking of the traditional shot-count metric. Extensive experiments demonstrate that LithoGRPO achieves state‑of‑the‑art performance over both optimization‑based and learning‑based methods, while maintaining efficient mask generation.

Deep Learning · Algorithms

Natalia Frumkin, Diana Marculescu

Text-to-image diffusion models remain computationally intensive: generating a single image typically requires dozens of passes through large transformer backbones (*e.g.*, SDXL uses ~50 evaluations of a 2.6B-parameter model). Few-step variants reduce the step count to 2–8, but still rely on large, full-precision backbones, making inference impractical on resource-constrained platforms, both on-device (latency/energy) and in data centers with multi-instance GPU (MIG) style GPU partitioning (limited memory/throughput per slice). Existing post-training quantization (PTQ) methods are further hampered by dependence on full-precision calibration. We introduce Q-Sched, a scheduler-level PTQ approach that adapts the diffusion sampler while keeping the quantized weights fixed. By adjusting the few-step sampling trajectory with quantization-aware preconditioning coefficients, Q-Sched matches or surpasses full-precision quality while delivering a $4\times$ reduction in model size and preserving a single reusable checkpoint across bit-widths. To learn these coefficients, we propose a reference-free Joint Alignment–Quality (JAQ) loss, which combines text–image compatibility with an image-quality objective for fine-grained control; JAQ requires only a handful of calibration prompts and avoids any full-precision inference during calibration. Empirically, Q-Sched yields substantial gains: a **15.5%** FID improvement over the FP16 4-step Latent Consistency Model and a **16.6%** improvement over the FP16 8-step Phased Consistency Model, demonstrating that quantization and few-step distillation are complementary for high-fidelity generation. A large-scale user study with **80,000** annotations further validates these results on both FLUX.1[schnell] and SDXL-Turbo. Code will be released.

Deep Learning · Large Language Models

Shengqin Wang, Wentao Yan, Huichi Zhou, Yihang Chen, Kun Shao, Zhizhong Zhang, Yuan Xie

Agentic multimodal models have garnered significant attention for their ability to leverage external tools to tackle complex tasks. However, it is observed that such agents often meet premature interaction collapse, caused by two primary reasons: 1) the terminal reward often appending on the last token prevents the advantage from distinguishing trajectories with exploratory behavior; 2) excessively redundant context hinders the agent from absorbing useful feedback. To address these issues, we propose the Deepening Reasoning MMSearchAgent, the framework leverages the structural proximity to derive advantage signals from the whole rollout trajectories in an entire batch, such that trajectories of different lengths are further encouraged to be generated, even when containing the same correct answer. Additionally, differentiated gaussian rewards are employed to dynamically calibrate interaction tolerance, thereby ensuring information reliability and reduce redundancy. To support multi-turn interaction training, we have constructed a multi-step deep-reasoning dataset including 3602 high-quality QA pair with at least 3 reasonning steps. Extensive experiments demonstrate that our method achieves state-of-the-art performance, outperforming the MMSearch-R1 by 8.4$\%$ on FVQA-test.

Theory · Online Learning and Bandits

Alhad Sethi, SOFIA SAGAR KAVALI, Shubhada Agrawal, Debabrota Basu, P. N. Karthik

We study one-sided and $\alpha$-correct sequential hypothesis testing for data generated by an ergodic Markov chain. The *null* hypothesis is that the unknown transition matrix belongs to a prescribed set $\cal P$ of stochastic matrices, and the *alternative* corresponds to a disjoint set $\cal Q$. We establish *a tight non-asymptotic instance-dependent lower bound* on the expected stopping time of any valid sequential test under the alternative. Our novel analysis improves the existing lower bounds, which are either asymptotic or provably sub-optimal in this setting. Our lower bound incorporates both the stationary distribution and the transition structure induced by the unknown Markov chain. We further propose an optimal test whose expected stopping time matches this lower bound asymptotically as $\alpha \to 0$. We illustrate the usefulness of our framework through applications to sequential detection of model misspecification in Markov Chain Monte Carlo and to testing structural properties, such as the linearity of transition dynamics, in Markov decision processes. Our findings yield a sharp and general characterization of optimal sequential testing procedures under Markovian dependence.

General Machine Learning · Evaluation

Chinh Hoang, Mohammad Hasan

Vision-language models (VLMs) generate fluent causal explanations for visual scenes, but does this fluency reflect genuine structural understanding? We address this question through a dual-probe methodology that isolates plausibility from faithfulness. The Text-Only Probe measures linguistic quality; the Chain-Text Probe requires models to first generate explicit causal chains before text responses. We define the Abstraction Gap (AG) metric as the normalized performance difference between probes, operationalizing the plausibility-faithfulness distinction from explainable AI research. Applying this methodology to eight VLMs using CAGE (Causal Abstraction Gap Evaluation), a benchmark of 49,500 questions across 5,500 images spanning Pearl's causal hierarchy, we find seven models exhibit AG exceeding 0.50: scoring 6--8 on text but below 2.5 on chains, often producing blank outputs. Fine-tuning on 45,000 chain-annotated examples fails to close the gap, indicating that explicit chain supervision cannot instill structural abstraction capability. Current VLMs optimize for plausible language without faithful structural understanding.

Theory · Deep Learning

Nathanaël Haas, François Gatine, Augustin Cosse, Zied Bouraoui

Understanding why gradient-based training in deep networks exhibits strong implicit bias remains challenging, in part because tractable singular-value dynamics are typically available only for balanced deep linear models. We propose an alternative route based on two theoretically grounded and empirically testable signatures of deep Jacobians: depth-induced exponential scaling of ordered singular values and strong spectral separation. Adopting a fixed-gates view of piecewise-linear networks, where Jacobians reduce to products of masked linear maps within a single activation region, we prove the existence of Lyapunov exponents governing the top singular values at initialization, give closed-form expressions in a tractable masked model, and quantify finite-depth corrections. We further show that sufficiently strong separation forces singular-vector alignment in matrix products, yielding an approximately shared singular basis for intermediate Jacobians. Together, these results motivate an approximation regime in which singular-value dynamics become effectively decoupled, mirroring classical balanced deep-linear analyses without requiring balancing. Experiments in fixed-gates settings validate the predicted scaling, alignment, and resulting dynamics, supporting a mechanistic account of emergent low-rank Jacobian structure as a driver of implicit bias.

Pingzhu Liu, Chunming He, Zunnan Xu, Chao Hao, Bo Zhao, Xingyu Shao, Jun Zhou, Zitong YU, Xiu Li

Unsupervised Camouflaged Object Detection (UCOD) aims to identify objects concealed in their surroundings without relying on pixel-level labels. Existing methods rely solely on simple post-processing of DINO high-dimensional features to generate pseudo labels for training. However, these methods suffer from two major limitations: 1) pseudo labels they easily generate contain excessive noise, causing the model to learn substantial incorrect information. 2) Although pseudo-label supervision allows the model to understand the task, it remains insufficient for generating fine-grained segmentation of the camouflaged objects. To address these issues, we propose DualUCOD, a novel UCOD method based on dual-branch contrastive learning that effectively detects camouflaged objects without pixel-level labels. Specifically, we propose the Dual-Eigenvector Spectral Pseudo-Labeling (DESPL) strategy, which fuses semantic and color cues into an affinity matrix. We then compute the eigenvectors of its normalized graph Laplacian and generate high-quality pseudo-labels using these eigenvectors. Furthermore, we introduce a Boundary-Guided Foreground-Background Refinement (BGFBR) module that explicitly incorporates boundary information to improve segmentation accuracy. Finally, we introduce a Dual-Branch Contrastive Learning (DBCL) module that constructs positive and negative pairs from the original and augmented images, aligning positive representations while contrasting them against negatives to enhance camouflaged object understanding. Extensive experiments demonstrate that DualUCOD outperforms state-of-the-art methods on different datasets in the unsupervised setting.